It must always be recognized that when the cells having rechargeable manganese dioxide cathodes, with zinc or other anodes as discussed above and noted hereafter, are assembled in their fully charged state. That is to way, the first cycle to which any such ell is subjected to, in use, is a discharge cycle, after which the cell is required to be re-charged for subsequent use. This is, of course, in contradistinction to nickel cadmium (Ni-Cd) cells, which must first be charged before they are capable of being used.
Because any cell in keeping with the present invention--whether a bobbin cell, or a coin or button cell--is subjected to discharge when it is first put into use, and because manganese dioxide cathodes (especially, in the presence of an alkaline electrolyte) have a tendency to swell during discharge, care must be taken to ensure that the cathode maintains its integrity and does not disintegrate or substantially swell in such a manner as to disturb the internal structure of the cell, thereby rendering it ineffective for further use. That, of course, is what happens or may happen generally if primary alkaline manganese dioxide-zinc cells are subjected to a charge cycle following discharge.
Generally, for cells in keeping with the present invention, the manganese dioxide cathode--which is specifically the subject of the present invention and which is discussed in greater detail hereafter--is provided together with an anode, a separator, and an electrolyte, all in ah suitable container, and sealed therein by a suitable closure. In general, the electrolyte is 6 N KOH to 12 N KOH. The anolyte--which is essentially electrolyte which is used for formulating the anode --is generally also 6 N KOH to 12 N KOH, but may have zinc oxide dissolved in it in an attempt to reduce corrosion of the zinc metal of the anode, and so as to provide an overcharge reserve for the cell.
The separator which is used between the cathode and the anode is generally a cellulose, non-woven material, which may optionally have a fibre structure in it or associated with it for re-inforcement.
When the anode is zinc, it is generally a zinc powder mixed with a gelling agent which may be such as NaCMC. Corrosion inhibitors such as mercury, lead, cadmium, indium, gallium, and thallium, may also be included in the anode formulation, in an attempt to reduce hydrogen gassing within the cell.
For a more complete understanding of the present invention, some discussion follows with respect to the characteristics of a zinc anode--as being the most typical anode used in commercial cells embodying a manganese dioxide cathode--and with respect to certain characteristics of manganese dioxide when used as a cathode.
First, having regard to a zinc anode as discussed above, it is noted that during the first few cycles of the rechargeable cell, a certain portion of the active zinc mass may become inactive. Typically, a gelled cylindrical anode has a central current collector (the nail) placed down its centre, particularly in such as cylindrical or bobbin-type rechargeable manganese-zinc cells, and the anode may have in the order of 50%-70% by weight (amalgamated) of zinc powder. Electrical conductivity within the gelled anode is established through the contact of the individual metallic zinc particles within the anode (the so-called zinc chain). However, as discharge of the cell proceeds, the highly conductive zinc particles are oxidized to become non-conductive ZnO or Zn(OH)2, each of which is a solid. Later, the zinc oxide or zinc hydroxide may dissolve to form zincate ions. However, after the electrolyte in the neighbourhood of the metallic zinc particles is locally saturated with zincate the compounds no longer dissolve and the discharge reaction will stop due to passivation of the anode. (This is particularly discussed in Falk and Salkind, "Alkaline Storage Batteries", published by John Wiley & Sons, 1969, at pages 156-159.)
When the cell is recharged, zinc is replated in the anode, initially near the nail or current collector, but the conductive zinc chain which originally existed can no longer be completely be re-established without a significant overcharge of the cell. The addition of conductive additives which do not participate in the discharge and charge reaction will remedy this situation, and is contemplated in a patent application assigned to the assignee herein, in the name of Kordesch, Sharma and Tomantschger, application Ser. No. 608,841 filed Nov. 5, 1990.
As to the manganese dioxide, Falk and Salkind (above) at pages 180-182, describe the discharge reaction of manganese dioxide in alkaline solution. The discharge reaction is quite complex, and may proceed in various steps. It is now generally accepted that the mechanism proposed by Kozawa best describes the discharge of manganese dioxide ("Batteries", Volume 1, Manganese Dioxide--edited by Kordesch--chapter 3). The MnO2 discharge curve has a sloping characteristic, indicating an homogeneous phase reaction. The potential of the MnO2 changes continuously while protons originating from the water of the electrolyte are introduced into the ionic lattice of the manganese dioxide, according to the equation; EQU MnO2+H2O+e=MnOOH+OH (Equation 1)
However, the MnO2 lattice expands, and at a certain point during the discharge, the mechanism changes. After that time the discharge occurs in a heterogeneous phase reaction, according to the reaction; EQU MnOOH+H2O+e=Mn(OH)2+OH (Equation 2)
This second reaction step involves the dissolusion of MnOOH in the form of (Mn(OH)4), with electrochemical reduction on the graphite in the manganese dioxide cathode to (Mn(OH)4)", and the precipitation of Mn(OH)2 from it.
Reference is made to a further copending application 400,712 assigned to the assignee hereof, in the name of Kordesch, Gsellmann, and Tomantschger, which discusses a practical approach to the problem of loss of capacity of manganese dioxide cathodes, now U.S. Pat. No. 5,011,752, issued Apr. 30, 1991. In that invention, it is proposed that the manganese dioxide material be pre-conditioned to have an oxidation state at the time that the cell is finally assembled and sealed between 1.7 and 1.9.
It should be noted that MnO2 which has been discharged in its homogeneous phase may be recharged. However, any Mn304 that is formed during discharge is not capable of being recharged. The Mn(OH)2 noted in equation 2 above, may be reoxidized to become Mn304. There is no evidence that MnOOH is to be found in a discharged manganese dioxide cathode.
The discharge of MnO2 is also discussed in, for example Dzieciuch et al U.S. Pat. No. 4,451,543 where it is suggested that MnO2 may be rechargeable to the two electron level. There, is was found that MnO2 was reduced in an homogenous phase to Mn01.6, thereby forming an alpha MnOOH (groutite) having a gamma structure. Beta MnO2 (chemical manganese dioxide--CMD) was only reduced homogenously to about MnO1.96 or MnO1.98.
Boden et al, J. Electrochem. Soc. 114, at 415 (1967) confirm that the discharge of EMD is an homogenous phase discharge, but they postulate an amorphous intermediate. This was because the internal resistance was found to rapidly increase with MnO1.6,and that it reaches a ten-fold value of MnO2 at about MnO1.4.
Euler, In Electrochimica Acta 15, at 1233 (1970) studied commercial battery electrodes, and revealed the influence of conductivity of the cathode mix and electrolyte penetration. This is complicated, however, by the ability of MnO2 cathodes to recuperate from an homogenous phase discharge. This suggests therefore, that there are potential gradients within the manganese dioxide cathode, under load conditions; and this suggests, therefore, that rechargeable MnO2 electrodes may have been locally over-discharged.
One further problem that develops generally in manganese dioxide cathodes is the possible migration of the zincate from the anode to the cathode. Zincate ions can be transported to the manganese dioxide cathode and there they form a mixed oxide, hetaerolite (ZnO.Mn203). The hetaerolite irreversibly affects the behaviour of a manganese dioxide cathode. It has been particularly recognized by Kordesch et al in Electrochmica Act 25 (1981) at 1495 to 1504 that the longevity of a rechargeable alkaline manganese dioxide cathode in its homogenous discharge phase above about MnO1.55 was limited by the mechanical failure of the electrode (cathode). It has been well shown that a manganese dioxide cathode expands during discharge and contracts during charge. Kordesch et al have shown that cycling an unconfined manganese dioxide cathode through four discharge-charge cycles resulted in the thickness of the MnO2 electrode becoming more than double its original thickness, and that the electrode failed due to the bulging and the mechanical disintegration which occurred. This was notwithstanding the fact that a binder (in this case, polysulfone) was employed.
Kordesch et al also demonstrated that if a similar electrode was confined by a perforated disc, the confined electrode continues its cycling life well beyond the fourth cycle; and that the change in dimension between the charged and the discharged electrode was only about half of that which occurred in the unconfined electrode. It was demonstrated that a mounting pressure of about 250-750 N/cm2 was required to increase the cycle life from less than about 5 cycles--shown, above, to be because of poor conductivity and mechanical disintegration and to achieve at least a cycle life of 75 cycles. A peak of 92 cycles was found at 500 N/cm2. However, it was also found that a higher mounting pressures the cycle life would drop because of the loss of pore volume within the manganese dioxide cathode, thereby creating problems with respect to electrolyte penetration within the cathode.
When the manganese dioxide cathode is present in the form of a sleeve or a disc, additional difficulties may arise. The internal resistance of the electrode may increase, and the mechanical disintegration of the cathode may be particularly severe. Kordesch, in "Batteries, Vol. 1" at pages 201 to 219 discusses these problems. Several prior art references show attempts to preclude the expansion of a manganese dioxide cathode during discharge and, indeed, to try to prevent its contraction during charge, including such matters as the addition of a binder such as a cement (U.S. Pat. No. 2,962,540); the addition of graphitized textile fibres (U.S. Pat. No. 2,977,401); the addition of latex binders (U.S. Pat. No. 3,113,050); use of combination binders such as cement and steel wool (U.S. Pat. No. 3,335,031); and the use of supplementing binders as described in U.S. Pat. No. 3,945,847. None of those patents, however, could preclude the mechanical disintegration of the cathode, apparently due to the limited binding strength of the materials being used.
Kordesch and Gsellmann in U.S. Pat. No. 4,384,029 teach cylindrical "bobbin" cells which may use mechanical enclosures such as tubes, springs, mechanical wedges, and perforated cylinders, to preclude expansion of the cathode during discharge of those bobbin cells. What this patent attempts to do is to create a constant volume cathode, which means that the cathode must always be under a certain mounting pressure at all times. That patent suggests that by increasing the mounting pressure, the number of usable cycles for the cell will increase. By providing the metal cage, which is essentially rigid, the tendency of the cathode to swell creates internal pressure within itself, which acts against the metal cage and between the cage and the can, thereby counteracting the tendency to swell; and by maintaining the cathode under pressure, it maintains a substantially constant volume during discharge as well as charge.
A different approach, using combinations of binders with a mechanical retainer or multiple mechanical retainers is disclosed in a further patent assigned to the present assignee, in the name of Kordesch, Gsellmann and Tomantschger, being U.S. Pat. No. 4,957,8267, issued Sept. 18, 1990.
What must be recognized is that, while it is shown that the use of means particularly such as the cages of the two Kordesch et al patents noted immediately above provides for a structure having up to several hundred cycles, there are several disadvantages to be considered. Particularly, where cement or other non-conductive binders are used, they may be preset in the range of typically 5%-10% by volume of the cathode, and therefore the amount of active ingredients that can be placed in the cathode is reduced. This results in a decrease in the usable capacity of the cell, and it may also result in a decrease int he conductivity of the cathode of the cell. On the other hand, if an insufficient amount of binder content is used, typically the manganese dioxide cathode may tend to crumble and/or crack, so that a coherent electrode is not achieved and its integrity is seriously affected.
If mechanical structures such as cages or screens are employed, then there is a significant increase in the material cost of the cell, as well as a significant increase in the cost of assembly of the cell. Indeed, there may be a significant effect and complication with respect to the use of high speed production equipment. Moreover, the use of a mechanical component such as a perforated iron cage or plate may significantly increase the probability of cell gassing within the cell.
Still further, the use of the mechanical cage or screen adjacent to the separator of the cell may significantly affect the capability of the cell to operate in a high drain condition. Any mechanical means which restricts the electrode interface between the cathode and the anode will act to limit the current density achievable within the cell.